IBM

IBM / awesome-agentic-workflow-optimization

Public

this is for fun, ain't it grand!

19
2
100% credibility
Found Mar 26, 2026 at 19 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
AI Summary

A curated list of research papers on optimizing workflows for large language model agents, categorized by static and dynamic techniques, accompanying an arXiv survey paper.

How It Works

1
🔍 Discover the Collection

You search online for ways to make AI helpers work more efficiently and stumble upon this handy list from IBM.

2
📖 Open the Guide

You click into the page and see a clear overview picture plus organized sections on different improvement methods.

3
📂 Explore Categories

You browse friendly sections like fixed plans or on-the-fly changes to find topics that match what you need.

4
📄 Pick Interesting Papers

You scan the tables of paper titles from recent years and click links to ones that spark your curiosity.

5
💡 Read and Learn

You dive into the papers, gaining ideas on smarter ways for AI teams to collaborate and solve problems.

6
✏️ Add Your Favorites

If you find a great paper missing, you easily suggest adding it following simple guidelines.

🎉 Master AI Workflows

Now you're equipped with the latest research to build faster, smarter AI assistants that get things done right.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 19 to 19 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is awesome-agentic-workflow-optimization?

This repo delivers a curated list of research papers on optimizing workflows for LLM agents, split into static (pre-defined templates) and dynamic (runtime tweaks) categories. It accompanies a survey paper on shifting from rigid setups to adaptive graphs in agentic systems, giving you quick links to 50+ recent works like DSPy and AutoFlow. Developers get a method-centric resource to scan agentic workflow papers without digging through arXiv.

Why is it gaining traction?

It stands out with tight organization—subsections like "in-execution editing" make spotting relevant agentic techniques instant, unlike scattered lists or broad LLM surveys. The IBM backing and arXiv survey tie-in hook researchers chasing efficiency in multi-agent setups, where pruning or topology evolution cuts token waste. Even with low stars, it's a no-fluff launchpad for "ain't it fun" experiments in agentic optimization.

Who should use this?

AI engineers building LLM agent pipelines for code gen or task orchestration, especially those debugging bloated multi-agent comms. Researchers prototyping dynamic workflows, like evolving topologies for Verilog or query-specific crews. Teams optimizing DSPy-style pipelines before scaling to production.

Verdict

Grab it as a free, focused paper index if agentic workflows are your jam—1.0% credibility reflects 19 stars and single-file simplicity, but solid categorization and fresh 2025 papers make it a constructive starting point over generic searches. Skip if you need code or benchmarks. (187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.